Design of Samples for International Migration Surveys: Methodological Considerations, Practical Constraints, and Lessons Learned from a Multi-Country Study in Africa and Europe
George Groenewold, Netherlands Interdisciplinary Demographic Institute (NIDI)
Richard Bilsborrow, University of North Carolina at Chapel Hill
The paper begins with a review of shortcomings of standard data sources to investigate the causes, consequences and mechanisms of international migration. It then argues that specialised household sample surveys are a powerful alternative. Specific sample design issues for the study of international migration in countries of origin and destination are then discussed, including the "rare elements" problem and approaches for dealing with it, and the tendency of migrant households to cluster in particular areas in countries of origin and destination. The paper then presents a model sample design approach permitting obtaining statistically representative results, and discusses its application in five migrant-sending countries (Egypt, Ghana, Morocco, Senegal, Turkey) and two migrant-receiving countries (Italy, Spain), covered by the NIDI/Eurostat study, "Push and Pull Factors of International Migration". Main design features, implementation constraints, and final approaches used are discussed, leading to a number of methodological recommendations.